dx.doi.org/10.

14227/DT290222P62

Update on Gastrointestinal Biorelevant Media and

Physiologically Relevant Dissolution Conditions

Daniela Amaral Silva1,2, Neal M. Davies1, Nadia Bou-Chacra2, Humberto Gomes Ferraz2, and Raimar
Löbenberg1*
1Faculty of Pharmacy and Pharmaceutical Sciences, University of Alberta, Edmonton, AB, Canada.
2Faculty of Pharmaceutical Sciences, University of Sao Paulo, Sao Paulo, Brazil.

email: [email protected]

ABSTRACT
Dissolution testing constitutes one of the most widely used in vitro performance tests during drug development and
routine quality control testing. It monitors the rate and extent of in vitro drug release (batch release test), and it is
often used to ensure consistent in vivo performance. The purpose of this review is to summarize and update the many
physiologically adapted media and buffers proposed over the years, focusing on the upper gastrointestinal tract because
this is where most drug absorption occurs. Emphasis will be given on the application of bicarbonate-based media
because this is the major buffering species in the human intestinal lumen. Due to the pragmatical difficulties of using
bicarbonate-based dissolution media, surrogate media with simpler buffer systems are desirable. Herein we describe
some of the proposed models and different approaches to develop such substitutes. Special consideration has to be
taken when dealing with enteric coated (delayed release) formulations because the interaction of coating polymer with
bicarbonate is very complex. All factors considered, using physiologically relevant conditions can ameliorate the risks
and enable drug development with increased likelihood to select formulations with the desired in vivo performance.

KEYWORDS: Bicarbonate buffer, dissolution, physiologically relevant, biorelevant, enteric coating

D
BACKGROUND At the time when USP apparatus 1 and 2 were introduced
issolution testing constitutes one of the most widely and FDA guidances were published, most of the
used in vitro performance tests during drug product molecules under development presented good aqueous
development and routine quality control testing. It solubility (BCS classes 1 and 3), and conventional dosage
monitors the rate and extent of in vitro drug release (batch forms (capsules and tablets) were the most common.
release test), and it is often used to ensure consistent Hence, establishing in vitro dissolution conditions with
in vivo performance (1, 2). The description of standard presumed in vivo relevance was reasonably simple (1,
dissolution apparatus by the United States Pharmacopeia 7). The development scenario has changed to molecular
(USP) in the 1970s together with guidance from the entities that are more potent accompanied with lower
United States Food and Drug Administration (FDA) in aqueous solubility (BCS classes 2 and 4). Although these
the late 90s propelled its broad application during the drug substances have enhanced many therapies by acting
various stages of drug development (1, 3). Alongside that, on new molecular targets, they also present significant
the introduction of the Biopharmaceutics Classification formulation and process development challenges,
System (BCS) in 1995 provided a simple but robust way to especially regarding the biopredictive power of previous
mechanistically describe the biopharmaceutical behavior traditional in vitro performance test methods (8).
of a drug (4). Under this system, drugs are classified based Hence, there was a need for advancement in the field of
on their solubility and permeability. These parameters dissolution testing (e.g., development of biorelevant and
may be used to predict the fraction of dose absorbed and physiologically relevant dissolution methods) to address
consequently its chances to become bioavailable (5, 6). the shortfalls of traditional methods.

* Corresponding author.

62 MAY 2022
www.dissolutiontech.com
Accordingly, the use of dissolution testing has expanded used for regulatory applications, meeting the specified
beyond the routine end-product release application to a criteria by the regulatory agency.
comprehensive analysis that can be implemented at the
various stages of the product life cycle (1, 5). Changes In vivo drug dissolution depends on the drug’s
in the regulatory landscape, such as the introduction of physicochemical properties and the GI fluid environment.
quality by design (QbD) concepts, have also contributed The current understanding of the human GI physiology
to the progression of dissolution methodology, linking allowed biorelevant dissolution media (BDM) to evolve,
quality tests to product performance in patients and facilitating the in vitro prediction of in vivo dissolution
ultimately therapeutic outcomes. Hence, there was a performance (12, 17–20). The many proposed BDM
push to develop dissolution media and apparatus that include various properties of the human GI tract, such
may mimic the human gastrointestinal (GI) tract to as pH, buffer species, buffer concentration, osmolality,
further understand in vivo dissolution mechanisms. The viscosity, surface tension, concentration, type of bile
innovation in this field has also evolved to the integration salts and lipolysis products, and physiological state, such
of in vitro dissolution data, applying different methods as fasted and fed states (21–24). Evidently conventional
and analytical techniques with modeling and simulation, dissolution media, such as simple USP buffers, fall short
and correlating with in vivo data (9–12). This approach in mimicking the properties and composition of GI fluids
is a robust way to select the best formulation with the (17). At the same time, the compendial methods that
desired in vivo performance. are approved by a given regulatory agency for products
marketed in that region cover different types of drug
The purpose of a particular dissolution test varies at the substances and dosage forms.
different stages of development (1). As first introduced by
Azarmi et al., there might be a need for more than one A more accurate prediction of the drug product’s in vivo
dissolution test for the same product (13). For example, performance is expected the closer the in vitro conditions
a quality control (QC) dissolution test is usually used to are to the in vivo environment. However, depending on
identify possible variations during product manufacturing the information one is seeking or on the physicochemical
and/or changes in product storage that could have an properties of the API (e.g., BCS class I), simulating all
impact on the product’s performance. This method aspects of the GI tract may or may not be necessary to
needs to be simple to be used in a typical routine QC evaluate the drug product performance. Based on this,
environment, such as conventional USP apparatus 1 or 2 Markopoulos et al. have suggested levels of simulation
and simple buffer media. At the same time, this method of luminal composition, as follows: Level 0 (pH); Level
has to demonstrate an appropriate level of discriminatory I (pH and buffer capacity); Level II (pH, buffer capacity,
power to confirm product consistency. On the other bile components, dietary lipids, lipid digestion products,
hand, a biorelevant/ physiologically relevant dissolution and osmolarity) and Level III (pH, buffer capacity, bile
method applies conditions that mimic the different components, dietary lipids, lipid digestion products,
physiological environments. These usually consist of non- osmolarity, proteins, enzymes, and viscosity effects) (25).
compendial media and apparatus, such as bicarbonate- The purpose of this review is to summarize and update
based buffers, biphasic dissolution to assess the impact of the many physiologically adapted media and buffers
concurrent drug absorption and multiple compartmental proposed over the years, focusing on the upper GI tract
apparatuses (14–16). This methodology is mostly used to because this is where most drug absorption occurs.
guide formulation selection and optimization. It typically Emphasis will be given on the application of bicarbonate-
starts during early development and may continue based media because this is the major buffering species in
through clinical testing and beyond. Lastly, a clinically the human intestinal lumen.
relevant dissolution method is any particular method in
which a link between in vitro dissolution data with in vivo PHYSIOLOGICALLY RELEVANT MEDIA
pharmacokinetic (PK) data can be established, creating Gastric Environment
an in vitro-in vivo correlation or relationship (IVIVC or The composition, pH, and surface tension are important
IVIVR), which is important for lifecycle management. The aspects to be considered when simulating the gastric
different methods used may or may not overlap with fluid (Table 1). The composition of the stomach fluid is
each other. However, they are useful during the research not merely hydrochloric acid (HCl); it also contains saliva,
and development stage where in vivo insight is desirable. digestive enzymes (pepsin and gastric lipase), food, and
The information retrieved from such methods can then refluxed fluids from the duodenum (26). The pH of gastric
be used to set specifications for the QC method to be fluids can vary greatly depending on the physiological

MAY 2022 63
www.dissolutiontech.com
 state (fed vs fasted), health-related conditions (such as (19). Even though the use of physiologically relevant
achlorhydria), and pharmacological treatments (such as surfactants is desirable to mimic in vivo conditions as
anti-acid agents). The reported pH range of gastric fluids closely as possible, these media can be unstable, difficult
is 1.5–1.9 under fasted conditions and 3.0–7.0 under fed to prepare, and costly. Hence, synthetic surfactants, such
conditions (the rate in which the pH changes is strongly as sodium lauryl sulphate (SLS) and polyethylene glycol
related to the type and size of the meal) (27–30). The tert-octylphenyl ether (Triton X 100) are often used as an
reported surface tension in gastric fluids ranges from alternative. These surfactants are added into compendial
30–46 mN/m (28, 31, 32). This could be indicative of the simulated gastric fluid without pepsin to form SGFSLS and
presence of surface-active agents, such as lecithin and SGFTriton, respectively (Table 1). This can be an interesting
lysolecithin (33). approach, but on the other hand it is important to be
aware that different types of surfactants can impact the
One of the earliest proposed media to simulate the product’s performance, leading to erroneous predictions
stomach in the fasted state was the artificial gastric of drug dissolution (19, 36).
fluid (AGF), described by Ruby at al. in 1996 (34). The
compendial simulated gastric fluid (SGF) and its version Another important aspect to consider is the difference
without pepsin (SGFsp) described in the USP presents a of fasted versus fed physiological states. Macheras et al.,
different composition than AGF, but a similar pH (pH 1.2), proposed the use of milk as a medium that can simulate
as shown in Table 1 (35). Many aspects of the gastric juice gastric components in the fed state because it contains
are addressed in these media, but qualities such as pH, similar ratios of fat, protein, and carbohydrates present
surface tension and pepsin concentration could be more in the western diet (37, 38). However, there are some
reflective of the in vivo values. drawbacks with the use of milk, such as batch- to-batch
variability in the milk composition (contributing to variable
To mimic in vivo conditions as closely as possible, Vertzoni dissolution data), the tendency of lipophilic compounds
et al. designed a fasted state simulating gastric fluid to bind to lipidic components of the milk, and the source
(FaSSGF) including compounds found in the intragastric of milk (goat vs. cow) (38).
environment, such as pepsin and sodium taurocholate

Table 1. Composition of Biorelevant Media to Simulate Gastric Fluid Under Fasted and Fed Conditions (12, 19, 34, 36)

SGF FeSSGF FeSSGF FeSSGF

AGF SGF SGFsp SGFSLS FaSSGF
Triton X Early Middle Late
Acetic acid 500 μL
- - - - - - 17.12 mM -
(for 1L)
Lactic acid (μL) 420
- - - - - - - -
(for 1L)
Lecithin (μM) - - - - - 20 - - -
Pepsin (g) 1.25 3.2 - - - 0.1 - - -
Sodium chloride (mM) - 34.22 34.22 34.22 34.22 34.22 148 237.02 122.6
Sodium citrate (mM) 2.34 - - - - - - - -
Sodium lauryl sulphate (mM) - - - 8.57 - - - - -
Sodium malate (mM) 2.81 - - - - - - - -
Sodium taurocholate (μM) - - - - - 80 - - -
Triton X 100 (mM) - - - - 1.55 - - - -
Sodium acetate (mM) - - - - - - - 29.75 -
Ortho-phosphoric acid (mM) - - - - - - - - 5.5
Sodium dihydrogen
- - - - - - - - 32
phosphate (mM)
Milk/buffer - - - - - - 1:0 1:1 1:3
Hydrochloric acid qs qs qs qs qs qs qs qs qs
pH 1.2 1.2 1.2 1.2 1.2 1.6 6.4 5 3
AGF: artificial gastric fluid; SGF: simulated gastric fluid; SGFsp: SGF without pepsin; SGFSLS: sodium lauryl sulphate; SGF TritonX: polyethylene glycol tert-
octylphenyl ether; FaSSGF: fasted state simulating gastric fluid; FeSSGF: fed state gastric fluid. Dash (-) indicates not applicable.

64 NOVEMBER 2018
Another approach was proposed by Jantratid et al. in Based on this, biorelevant media, e.g., USP simulated
2008 (12). The authors proposed a “snapshot” approach intestinal fluids, were developed to simulate the pH
to capture the changes in the composition of the gastric including components present in the human GI tract,
fluid associated with digestion and gastric emptying such as bile salts and lecithin. Osmolality, pH, and surface
process (12). Table 1 describes the composition of early, tension were adjusted to physiological values. According
middle, and late fed state gastric environments (FeSSGF). to FDA guidance and other sources, simulated intestinal
The early stage media corresponds to the first 75 min fluid with pancreatin (USP-SIF) and without enzyme (SIF-
after meal ingestion, the middle stage to 75–165 min, and blank) reflect the physiologic conditions of the small
the late stage is beyond 165 min. intestine better than other simpler buffer systems (16,
40, 41).
Small Intestinal Environment
Biorelevant Dissolution Media Another example of biorelevant media is the fasted
The composition of biorelevant dissolution media for and fed simulated intestinal fluid (FaSSIF and FeSSIF)
simulating human small intestine fluids are presented in proposed by Dressman in 1998 and its many adaptations
Table 2. The bicarbonate ions secreted into the intestinal (30). The human intestinal lumen is buffer by bicarbonate;
lumen neutralize the gastric fluid that is emptied in however, due to pragmatical reasons, other buffers are
the intestines. The reported pH range under fasted typically used to mimic the physiological pH of intestinal
conditions is 5.8–6.5 in the duodenum, 5.3–8.1 in the fluids (30). For example, FeSSIF uses acetate buffer to
jejunum, and 6.8–8.0 in the ileum (31). Bile salts are also adjust the pH to 5.0. Moreover, the prevalent bile salt in
secreted into the intestines, and the formation of micelles the human bile is cholic acid, but sodium taurocholate
results in a much lower surface tension compared to the (conjugate of cholic acid with taurine) was chosen to be
gastric fluids. The surface tension of the intestinal fluids the most representative bile salt in vitro. Biorelevant
is even lower under fed conditions due to the higher media contain bile salts and phospholipids, and when
concentration of bile (39).
Table 2. Composition of Biorelevant Media to Simulate Human Small Intestine Fluids in the Fasted and Fed State and Canine Gastric and
Intestinal Biorelevant Media (5, 12, 30, 21, 54)
Media pH Components
USP SIF 6.8 NaOH (qs pH); KH2PO4 (6.8 g); Pancreatin (10.0 g); Deionized water qs 1L
FaSSIF 6.5 NaOH (qs pH); KCl (103.29 mM); Bile salt (Sodium taurocholate) (3 mM); Phospholipid (lecithin) (0.75 mM); Potassium
dihydrogen orthophosphate (28.66 mM); Deionized water qs 1L
FeSSIF 5.0 NaOH (qs pH); KCl (203.89 mM); Bile salt (Sodium taurocholate) (15 mM); Phospholipid (lecithin) (3.75 mM); Acetic acid
(144.05 mM); Deionized water qs 1L
FaSSIF V2 6.5 NaOH (34.8 mM); NaCl (68.62 mM); Bile salt (Sodium taurocholate) (3 mM); Phospholipid (lecithin) (0.2 mM); Maleic acid
(19.12 mM); Deionized water qs 1L
FeSSIF V2 5.8 NaOH (81.65 mM); NaCl (125.5 mM); Bile salt (Sodium taurocholate) (10 mM); Phospholipid (lecithin) (2 mM); Maleic acid
(55.02 mM); Glyceryl monooleate (5 mM); Sodium oleate (0.8 mM); Deionized water qs 1L
FeSSIF Early 6.5 NaOH (52.5 mM); NaCl (145.2 mM); Bile salt (Sodium taurocholate) (10 mM); Phospholipid (lecithin) (3 mM); Maleic acid (28.6
mM); Glyceryl monooleate (6.5 mM); Sodium oleate (40 mM); Deionized water qs 1L
FeSSIF 5.8 NaOH (65.3 mM); NaCl (125.8 mM); Bile salt (Sodium taurocholate) (7.5 mM); Phospholipid (lecithin) (2 mM); Maleic acid (44
Middle mM); Glyceryl monooleate (5 mM); Sodium oleate (30 mM); Deionized water qs 1L
FeSSIF Late 5.4 NaOH (72 mM); NaCl (51 mM); Bile salt (Sodium taurocholate) (4.5 mM); Phospholipid (lecithin) (0.5 mM); Maleic acid (55.09
mM); Glyceryl monooleate (1 mM); Sodium oleate (0.8 mM); Deionized water qs 1L
SEIF 6.5 NaN3 (6 mM); NaCl (98 mM); Bile salts* (4 mM); Phospholipid (Lyso-phosphatidylcholine) (1 mM); Cholesterol (0.25 mM);
Sodium dihydrogen phosphate (18 mM); Sodium hydrogen phosphate (12mM)
FaSSGFc I 1.2–2.5 HCl (~3.6 - 82 mM); NaCl (14.5 mM);Sodium taurocholate (0.1 mM); Sodium taurodeoxycholate (0.1 mM); Lecithin (0.025
mM); Lysolecithin (0.025 mM); Sodium oleate (0.025 mM)
FaSSGFc II 2.5–6.5 NaOH (~14.5 - 40 mM); NaCl (18.81 mM); Sodium taurocholate (0.1 mM); Sodium taurodeoxycholate (0.1 mM); Lecithin
(0.025 mM); Lysolecithin (0.025 mM); Sodium oleate (0.025 mM); Maleic acid (21.68 mM)
FaSSIFc 7.5 NaOH (21.66 mM); NaCl (59.63 mM); Sodium taurocholate (5.0 mM); Sodium taurodeoxycholate (5.0 mM); Lecithin (1.25
mM); Lysolecithin (1.25 mM); Sodium oleate (1.25 mM); Sodium dihydrogen phosphate (28.65 mM)
*Sodium salts of the following conjugates: glycocholate (1 mM), glycodeoxycholate (0.7 mM), glycochenodeoxycholate (1 mM), taurocholate (0.5 mM),
taurodeoxycholate (0.3 mM), taurochenodeoxycholate (0.5 mM).
SIF: simulated intestinal fluid with pancreatin; FaSSIF: fasted simulated intestinal fluid; FeSSIF: fed simulated intestinal fluid: V2: version 2; SEIF: simulated
endogenous intestinal fluid; FaSSGFc: canine fasted-state simulated gastric fluid; FaSSIFc: canine fasted-state simulated intestinal fluid.

MAY 2022 65
www.dissolutiontech.com
 simulating the fed state, also monoglycerides and free of the investigated active pharmaceutical ingredient
fatty acids. The composition of FaSSIF and FeSSIF are (API), whereas the sodium dodecyl sulfate (SDS) solutions
given in Table 2. overestimated solubility. Finally, the proposed FaSSIF-V3
composition contained glycocholate and taurocholate
Revised versions of FaSSIF and FeSSIF (FaSSIF-V2 and with 0.2 mM cholesterol (18).
FeSSIF-V2, respectively) were developed to address some
of the shortcomings of the initially proposed media. Cristofoletti and Dressman used FaSSIF-V3 with reduced
For example, Persson et al. reported that cyclosporine, phosphate buffer concentration (5.0 mM) (45, 46). The
danazol, griseofulvin, and felodipine presented between rationale behind this approach was to use a buffer system
2–5-times higher solubility values in fed human intestinal that would match the pH at the particle’s surface, utilizing
fluid (HIF) compared to FeSSIF (42). This could be due physiologically relevant bicarbonate buffer (BCB). For
to the lack of neutral lipids in the FeSSIF composition. this purpose, ibuprofen was used as the model drug. The
Additionally, the purity of bile salts can also have an authors reported that the proposed 5.0-mM phosphate
impact on the solubility of poorly soluble drugs. Wei buffer FaSSIF-V3 was able to predict in vivo differences in
and Löbenberg reported the solubility of glyburide in peak and extent of exposure between test and reference
biorelevant media with crude bile salts to be over 2-fold ibuprofen formulations (46).
higher than when pure bile salts were used in FaSSIF (43).
Additionally, the reported in vivo bile salt concentration When analyzing the fed state, as shown in Table 2, the
is lower than the concentration used previously (12, 32). main differences between FeSSIF and FeSSIF-V2 are the
concentrations of bile salts and lecithin, the replacement
Psachoulias et al. proposed a method to predict the of phosphate for maleate buffer resulting in lower
concentration and potential precipitation of lipophilic osmolality and buffer capacity values, and the addition
weak bases using an upgraded version of FaSSIF-V2 of glyceryl monooleate and sodium oleate to reflect the
(FaSSIF-V2plus) (44). The proposed in vitro method was presence of lipolysis products (32).
composed of a gastric and duodenal compartment along
with a reservoir. FaSSIF-V2plus was used in the duodenal Similarly to SGF, Jantratid and Dressman developed a
compartment. The composition of FaSSIF-V2plus is similar snapshot media to simulate the intestinal fluids in the
to FaSSIF-V2; in addition to all FaSSIF-V2 components, the fed state. The authors proposed the inclusion of lipolysis
“plus” version contains free fatty acid (sodium oleate, 0.5 products and changes in parameters such as bile salts
mM) and cholesterol (0.2 mM). The authors concluded concentration, osmolality, buffer capacity, and fluid pH
that for some weak bases, such as ketoconazole, FaSSIF- according to the early, medium, and late stages after food
V2plus is a superior fluid for investigating the drug’s intake (12).
intraluminal precipitation. Biorelevant media have been shown to be very useful in
Later, Fuchs et al. further proposed an updated version assessing the in vivo solubility of compounds. Söderlind
of the fasted state biorelevant media based on the up to et al. studied the solubility of 24 molecules in FaSSIF,
date physiological composition of fasted HIF at that time FaSSIF-V2, and HIF. FaSSIF-V2 solubilities correlated better
(18). The proposed media was named FaSSIF-V3. The with solubilities in HIF for neutral compounds, whereas
surface tension was considered as a surrogate parameter for acidic and basic compounds the solubility in FaSSIF and
in establishing the medium’s correctness. Several FaSSIF-V2 were similar (47). A similar trend was observed
prototypes were investigated containing five different bile by Fagerberg et al. (48). The authors reported that the
salts (taurocholate, glycocholate, tauroursodeoxycholate, estimation of the in vivo solubility of poorly soluble
taurochenodeoxycholate, and glycochenodeoxycholate), compounds was more accurate in biorelevant media. This
as well as replacing lecithin with its hydrolysis products was particularly true for bases and neutral molecules,
(lysolecithin and sodium oleate). Additionally, a mixture which display higher solubility in FeSSIF compared to
of glycocholate and taurocholate, with or without 0.2 mM FaSSIF. The opposite was observed for acidic drugs (48).
cholesterol, were investigated. The authors assessed the Biorelevant media have also been widely used to forecast
solubility of 10 model compounds and observed that the the in vivo performance drugs, achieving good IVIVC in
amount and type of phospholipids and bile salt significantly some cases, but not always (49–53). Other biorelevant
impacted the solubility and surface tension in the various media have also been proposed to simulate fluids in
prototypes. Additionally, the authors reported that blank the fasted state small intestine, such as the simulated
buffers tend to underestimate the physiological solubility endogenous intestinal fluid (SEIF), described by Kossena

66 MAY 2022
www.dissolutiontech.com
et al. (Table 2) (54). Since the focus of this review is on the dissolution media poorly representative of the in vivo
upper GI tract, colonic fluids are not included in Table 2. environment, failing to reflect in vivo characteristics
such as ionic strength, buffer capacity, fluid volume, and
The use of bicarbonate-based biorelevant media has been viscosity (16).
proposed in the literature (55, 56). Litou and colleagues
assessed a level II biorelevant media based on BCB to The pH along the GI tract is maintained by bicarbonate
simulate the contents of upper small intestine under ions, which are present in pancreatic, hepatic, and
conditions of reduced acid secretion in the stomach intestinal secretions (58, 59). Hence, the development
(56). The authors reported that bicarbonates were not of suitable in vitro dissolution media based on BCB has
important in estimating drug precipitation and that level gained much attention because it closely mimics the
II biorelevant media underestimated the concentration environment of the intestinal fluids and can thus improve
of the given compounds in intestinal human aspirates; in vitro-in vivo correlations compared to phosphate
however, more data are needed to confirm this finding as buffers (60).
the usefulness of bicarbonate in biorelevant dissolution
testing may be compound specific (56). For example, Jede In vivo, the pH is held stable by the constant supply of
et al. investigated the supersaturation and precipitation bicarbonate-containing secretions in the intestines.
kinetics of weak bases using a transfer model with On the other hand, the application of BCB as an in vitro
biorelevant BCB (55). The authors compared bicarbonate- dissolution medium is challenging due to the evaporation
and phosphate-based FaSSIF and found that bicarbonate- of CO2(g) from the aqueous phase causing the pH to rise.
based FaSSIF had better predictive power compared This can lead to changes in the buffer strength and poor
to phosphate-based FaSSIF. They concluded that the reproducibility of the dissolution test. Hence, the first
proposed model is a promising approach to increase the step in establishing a stable BCB is to maintain CO2(aq) and
predictive power of in vitro tests, thus contributing to a CO2(g) at equilibrium (Eq. 1).
more biorelevant drug development process (55).
CO2(g)
Even though biorelevant media have been extensively ↑↓
used, its preparation can be time-consuming, costly, and H2O(l) + CO2 (aq) ← (1)
→ H2CO3 (aq) ←
→ H (aq) + HCO3 (aq)
+ -
it may present a short-shelf life for utility. Furthermore,
the buffering species in the human intestinal lumen is
bicarbonate, whereas FaSSIF uses phosphate, FeSSIF uses One of the ways to stabilize the BCB pH is to purge the
acetate, FaSSIF-V2 and FeSSIF-V2 use maleate. Simpler medium with CO2 gas, thus supplying CO2(g), which
and more physiologically relevant dissolution media are compensates its loss from the aqueous medium.
therefore desired. Automated systems have been developed to adjust the
pH by sparging gas according to the pH shift and were
Moreover, in addition to FaSSGF, FeSSGF, FaSSIF-V2, reviewed by Amaral Silva et al. and others (16). However,
and FeSSIF-V2, there are several biorelevant media bubbling gases into the dissolution medium can be
described in the USP general chapter <1236> Solubility problematic due to the hydrodynamic disturbances in the
Measurements, such as human simulated colonic fluid— dissolution vessel, which can affect the dissolution rate of
proximal colon; human simulated colonic fluid—distal certain drugs and lead to failure in meeting compendial
colon, canine fasted-state simulated gastric fluid (FaSSGFc requirements. Another concern is the possible foaming
pH 1.2–2.5); canine fasted-state simulated gastric fluid when surfactant-containing media are used.
(FaSSGFc pH 2.5–6.5); canine fasted-state simulated
intestinal fluid (FaSSIFc), and bovine simulated ruminal Preventing the escape of CO2 instead of purging the
fluids. Biorelevant canine media are described in Table medium has been proposed as an alternative to control
2. For the composition of the other media, refer to USP the medium pH. Approaches such as sealing the
general chapter <1236> Solubility Measurements. dissolution vessel or using a liquid paraffin layer on top of
the dissolution medium have been effective in stabilizing
Physiologically Relevant Dissolution Media –
the media pH (61, 62). Nevertheless, these were closed
Bicarbonate Buffer (BCB)
systems, so dynamic pH regulation was not possible.
At present, the most widely applied dissolution media
are phosphate-based buffers (16, 57). However, the To circumvent this, Scott and colleagues have recently
concentration of phosphates in the intestinal luminal studied the use of a novel bicarbonate-based dissolution
fluids is insignificant. This makes phosphate-based system that supplies N2 (pH increasing) and CO2 (pH

MAY 2022 67
www.dissolutiontech.com
 decreasing) gases above the dissolution medium without of a pH-increasing gas (e.g., N2 or He) has an indirect
purging into the solution (see schematic of the device in effect in increasing the medium pH (65, 66). As the pH-
ref 60). The system is composed of an enclosure device increasing gas is supplied, the partial pressure of CO2 is
with two inlets that supply N2 and CO2. The gases are reduced, decreasing the dissolved CO2(aq) in the medium,
distributed through a ring-shaped diffuser and released thus increasing the pH (equilibrium in Eq. 1 shifts to the
through outlets pointing towards the surface of the left). Scott et al. observed that the CO2 supply was much
dissolution medium. The authors report that this method more efficient in decreasing the medium pH than N2 was
regulated the pH of the BCB without substantial disruption in increasing the pH (60). This is most likely due to the
to the surface of the media and that no foaming was indirect effect of N2, thus taking longer for the pH change
observed when surfactant-containing medium was used. to be observed.

The approach taken by Scott et al. is similar to the one BCB Kinetics and Mathematical Models
reported by Boni et al., in which the CO2(g) was supplied Although BCB is physiologically relevant, its application
above the medium to maintain the pH throughout the has been limited because of the pragmatical hurdles,
dissolution test (57, 60). However, the setup proposed and some authors have doomed it as a medium with
by Boni et al. was not effective because the dissolution restricted suitability for dissolution testing (57). Matching
vessel only had a conventional lid (open system) that did the effective buffering pKa of bicarbonate at the solid-
not prevent the escape of the supplied gas. Hence, the liquid surface (diffusion layer) of a dissolving solid with a
enclosure method is a superior design in the sense that it surrogate buffer system is a way to simplify the dissolution
prevents gas escape, thus improving the efficiency of gas conditions while maintaining physiologic relevance in
supply. The authors concluded that this novel system is terms of buffering capacity at the diffusion layer (67, 68).
a step towards the application of physiologically relevant
BCBs as dissolution media that meets compendial When the whole system is at equilibrium (Eq. 1), the pKa
requirements. of the BCB system is 6.04, which is the situation in the
bulk solution in a dissolution vessel (16, 67). However,
Sakamoto et al. proposed a simple and facile method that in the diffusion layer around dissolving solutes the
allows the use of BCB for dissolution testing (see schematic interconversion, H2O(l) + CO2(aq) ← → H2CO3(aq), does not
of the device in ref. 63). The authors developed a floating equilibrate very rapidly compared to the fast diffusion
lid system that prevents the escape of CO2 from the BCB processes. Therefore, BCB behaves as having an effective
solution. The lid is made of a 5-mm-thick styrofoam that pKa in the diffusion layer that is different from that in
covers the surface of the medium almost completely but bulk. This value is lower than 6.04 (bulk), but higher than
not in a tight-sealing configuration. The buffer is added to the intrinsic pKa of 3.30 (H2CO3(aq) ←→ H (aq) + HCO 3(aq)).
+ −

the dissolution vessel and the lid is placed on top of it. The As a result, the ability of BCB in buffering the diffusion
medium pH was adjusted by adding HCl via a small hole. layer against incoming ionizable solute is weakened and
The authors investigated the suitability of this method for the in vivo dissolution rate is slower compared to highly
a 6.0–7.5 pH range and 2–50-mM BCB concentration. In concentrated compendial buffers.
all cases, the pH change was less than 0.1 pH unit after
3.5 h when the floating lid method was used, whereas Based on this, investigators have proposed the reduction
without the lid, the pH increased by more than 1 pH unit in molarity of nonbicarbonate-based surrogate buffers
within 3.5 h. The authors concluded that the floating lid as a possible approach to increase its biopredictability,
method would be useful for formulation development thus matching the typically slower in vivo dissolution (15,
while covering the physiological intestinal and colonic 67, 68). For example, Tsume et al. showed that ibuprofen
conditions in terms of pH and buffer concentration. tablets had slower in vitro dissolution in phosphate 10 mM
compared to 50 mM at a starting pH of 6.0 (69). This can
It is interesting to note that the pH of the medium can be explained by Mooney’s stagnant film-based dissolution
be adjusted either by adding HCl/NaOH or by sparging model, i.e., more diluted buffers have a reduced buffer
gases or a combination of both (60, 63, 64). In the case capacity, which translates into a lower ability to counter
of sparging, when CO2 gas is supplied and diffused into the acidifying effect of the dissolving ibuprofen at the
the medium the CO2(aq) interacts with water, generating diffusion layer pH (70). In highly concentrated buffer
carbonic acid, which in turn dissociates, releasing systems, such as compendial buffers, an abundance
hydrogen ion, culminating in the pH decrease (equilibrium of the buffer’s conjugate base species surrounds the
shown in Eq. 1 shifts to the right). Reversely, the sparging drug particle. This leads to prompt neutralization in the

68 MAY 2022
www.dissolutiontech.com
diffusion layer, i.e., the buffer species readily consumes BCB not behaving as having a pKa exceeding 6 in terms
the ions formed on the dissolving drug surface. Hence, of promoting the dissolution of ionizable solids. The
the pH in the diffusion layer is similar to the bulk, resulting RNE model has been shown to successfully estimate
in a higher dissolution rate (15, 45, 71, 72). Conversely, the pH on the surface of a solid particle in BCB, and the
when the buffer system is less concentrated (as in vivo), Mooney model can be used to estimate the phosphate
the neutralization is slower. concentration that would give the same surface pH (pH0)
(67, 74). Thus, a proper surrogate buffer molarity can
Different models have been proposed to predict the drug be used that would give good matches to physiological
flux, thus enabling calculation of the surrogate buffer bicarbonate in terms of drug dissolution. This shows that
molarity to determine a good match to physiological in some cases it is feasible to develop surrogate buffers
bicarbonate in terms of drug dissolution. This includes but for bicarbonate.
is not limited to the equilibrium model (which assumes
that H2CO3 and CO2 are at equilibrium), the carbonic Furthermore, Salehi et. al. incorporated into the RNE
acid ionization (CAI) model (hypothetical situation model other properties such as medium hydrodynamic
where neither hydration or dehydration is assumed), effect and drug particle size distribution (75). The authors
the irreversible reaction (IRR) transport model, and the described it as a hierarchical mass transfer (HMT) model
reversible non-equilibrium (RNE) model. that considers drug properties (intrinsic solubility,
acid/ base character, pKa, particle size, and particle
Krieg et. al. proposed the IRR transport model to polydispersity) as well as GI fluid properties and fluid
develop more physiologically relevant buffer systems hydrodynamics (bulk pH, buffer species concentration,
for dissolution testing (73). This model assumes the fluid shear rate, and convection).
dehydration process (H2CO3(aq) → H2O(l) + CO2(aq))
is an irreversible chemical reaction because it is The findings reported by Álvarez et al. further reinforce
approximately 500 times faster than the hydration rate. that the current compendial buffers concentrations
This approximate model yielded improved predictions for seem to be too high to correlate with the in vivo
the intrinsic dissolution rates of ibuprofen, ketoprofen, carbonate concentration (76). The authors investigated
and indomethacin in BCBs. However, Al-Gousous et the in vitro dissolution of ibuprofen tablets in different
al showed that this assumption was not as accurate pharmacopeial media at both 50 and 75 rpm. The media
(74). They proposed the RNE model, which does not investigated by the group included 130-mM HCl pH 1.2,
make any equilibrium assumptions. It not only includes 540-mM acetate buffer pH 4.5, and 70-mM phosphate
both the hydration and dehydration rates (H2O(l) + buffer pH 6.8. In all media, the dissolution profiles
CO2(aq) H2CO3(aq) →) but also accounts for the fluxes of all
← showed similarity at both rotation speeds. However, the
species involved in the mass transfer process. The RNE in vivo bioequivalence studies revealed that only one
model predicted the flux values obtained in the intrinsic out of the three test formulations was bioequivalent to
dissolution experiments more accurately compared to the reference. Hence, these in vitro tests were not able
the other models (74). to detect differences regarding the rate of absorption.
Based on this finding, the authors concluded that there
It is of crucial importance to understand the kinetics of remains a need to develop dissolution conditions that can
bicarbonate at the diffusion layer of a dissolving particle. predict bioequivalence outcomes and that the application
For example, in the equilibrium model, BCB would have a of biowaivers to BCS class IIa drugs would not be feasible.
pKa close to the bulk pH, resulting in effective buffering
at the surface of the dissolving drugs (overestimation). In contrast, Hofmann et. al. studied the dissolution of
In the CAI model, the assumption that hydration and ibuprofen in physiologically relevant BCB and reported that
dehydration reactions do not happen would mean that the in vitro dissolution profiles in bicarbonate compared
the buffer pKa would be much lower than the bulk pH, reasonably well with the in vivo intestinal dissolution of
resulting in a very poor ability to buffer the surface of the tested suspensions (67). They concluded that this
the dissolving drug (underestimation). Similarly, the IRR demonstrates the possible potential toward extending
transport model would also underestimate the drug biowaivers to BCS class IIa compounds.
flux, but not to the extent as of the CAI model because
it includes an irreversible dehydration reaction. The RNE Amaral Silva et. al. tested a 5-mM phosphate buffer as the
model represents an intermediate situation in which surrogate buffer for ibuprofen based on the IRR model
the reactions occur but do not reach equilibrium. In this described by Krieg et. al (15). The authors also observed
case, as previously mentioned, this situation results in a slower dissolution rate of ibuprofen immediate-release

MAY 2022 69
www.dissolutiontech.com
 tablets in low buffer capacity (5 mM) compared to prednisolone microparticles, pellets, and tablets (60).
compendial buffer (50 mM) and that compendial buffer They observed that in phosphate buffer the drug release
lacked discriminatory power (15, 67, 76). The authors was immediate after the 2-h acid exposure for the all
pointed out that the rapid in vitro dissolution rate cannot the tested dosage forms with no significant difference
be translated to the observed in vivo dissolution rate of among the dissolution profiles. On the other hand, in
ibuprofen. In contrast from the methodology used by BCB there was a long lag time for the onset of drug
Álvarez at al., in which different absorption rates could not release. An interesting observation highlighted by the
be detected in vitro, Amaral Silva and colleagues utilized authors was a shorter lag time for the microparticle
the low-capacity surrogate buffer in a biphasic dissolution formulation compared to pellets and tablets, which could
system (76). This system is composed of an organic layer be explained by the larger surface area available for
on top of the aqueous medium, thus mimicking the polymer dissolution. Similarly, Sakamoto et al. reported a
concurrent in vivo processes of drug dissolution and 30-min disintegration time and similar release profiles for
absorption. The addition of the organic phase works as EC 5-ASA tablets in a phosphate-based buffer, whereas in
a sink to the aqueous layer, assisting the medium pH BCB the disintegration time was about 4–8 h, with large
maintenance by the removal of the dissolved drug from variation (63).
the aqueous medium. Hence, the pH changes that are
expected when a low buffer capacity medium is used are With this in mind, the most biorelevant dissolution media
reduced. This is a valuable approach to investigate the for EC formulations would be a bicarbonate-based one.
drug product performance with improved physiological As highlighted before, the routine use of BCB is technically
relevance (15). difficult and even unfeasible for disintegration testing and
dissolution apparatuses, such as reciprocal cylinder (78).
Based on this, we herein suggest the use of a biphasic Therefore, similarly to small drug molecules, developing
system with the aqueous layer composed of BCB. a non-volatile surrogate buffer for EC products is of great
Adding paraffin on top of the buffer has been previously interest. However, EC polymers, being poly-acids with
proposed to prevent the CO2 escape, but drugs do not ionizable carboxylic groups, are much more complex than
typically partition to the liquid paraffin layer (61). We small molecules as its dissolution includes different phases
believe that the use of BCB coupled with an organic layer (31, 77, 79–81). In an environment with low pH values
(octanol) would not only prevent the escape of CO2 – (such as the stomach) the carboxyl groups are not ionized,
thus taking away the need to sparge the medium – but so the polymer is insoluble, resisting disintegration and
it would also allow assessment of the drug partitioning dissolution which prevents drug release. When the EC
(absorption). This would be a very robust physiologically dosage form is exposed to the intestinal fluids (higher
relevant approach and we suggest that future in vitro pH and buffered by bicarbonate) and when the pH0
studies along this line be conducted. (surface pH) of the polymer is above its pKa (dissolution
pH threshold), its ionization is promoted (77). Due to
Application to Enteric-Coated Formulations and Design
electrostatic repulsion, the polymer relaxes, swells,
of Surrogate Buffers
and undergoes chain disentanglement allowing further
Oversimplification of the dissolution conditions, for
ionization of other polymer chains, which diffuse away to
example using a surrogate buffer instead of BCB, may
the bulk solution (79, 81). This consists of the dissolution
not be relevant or proper for certain formulations. This is
phases of pH-responsive polymers, ultimately leading to
the case for delayed release drug products. Formulations
disintegration and dissolution of the dosage form.
coated with pH-responsive polymers have been shown to
have poor in vivo performance (77). One of the reasons Recently, Blechar et al. proposed a mechanistic approach
for this is the lack of biopredictability of the buffers used to enable the development of surrogate buffers for EC
for in vitro performance testing, preventing suitable in products with little bench work. As described before,
vitro product evaluation (78). The great discrepancy in the effective pKa of BCB in the diffusion layer (pKaeff)
the performance of delayed release (enteric-coated [EC]) is different from other buffers such as phosphate and
products in physiologically relevant BCB vs. phosphate maleate (pKa of 6.8 and 5.8, respectively) and different
buffer is well recognized in the literature, as highlighted from the bulk where everything is at equilibrium (78). For
by Amaral Silva et al. (77). This performance problem small molecules under regular hydrodynamic conditions
persists today, and recent reports by Scott et al. and the pKaeff of bicarbonate lies between 4 and 5 (78, 82).
Sakamoto et al. have corroborated these findings (60, 63). However, the complex behavior of EC polymers makes it
difficult for a direct calculation.
Scott and colleagues investigated the release of EC
70 MAY 2022
www.dissolutiontech.com
Besides the diffusion layer, a viscoelastic gel layer is bioequivalent. Solely satisfying the in vitro standard
formed on a polymer’s surface (Fig. 1), as opposed to for drug dissolution does not guarantee similar in vivo
only a diffusion layer on a particle’s surface. The gel layer behavior. On the other hand, when these formulations
presents an increased diffusional resistance which reduces were tested in BCB, a great discrepancy was observed,
the diffusion rate of the buffer species. Consequently, where the test formulation had a much more delayed
the time available for the interconversion between CO2 onset of dissolution than the reference. The use of non-
and H2CO3 is increased, allowing the equation, H2O(l) + physiologically relevant dissolution media during the drug
CO2(aq) ←
→ H2CO3(aq), to approach equilibrium. As a result, product development phase can be misleading, causing
the pKa of bicarbonate in the gel layer is increased poor selection of prototype formulations. Therefore,
compared to the pKaeff in the diffusion layer. Finally, both it was further evidenced that using BCB can de-risk the
the pKaeff (diffusion layer) and higher pKa in the gel layer development of generic EC formulations, increasing the
will control the polymer’s surface pH. Therefore, the gel likelihood for a successful BE.
layer increases the effective interfacial buffering pKa of
bicarbonate.

Figure 1. Representation of the solid-liquid interface of a dissolving

enteric polymer. From the gel layer to the bulk solution, the pH increases
and the viscosity decreases. N: neutral (unionized carboxylic acid, -COOH);
Minus (-) indicates negative charge (ionized carboxylic acid, -COO-).
Adapted from Blechar et al (79) under the Creative Commons Attribution
License (https://creativecommons.org/licenses/by/4.0/).

The authors performed dissolution experiments in Figure 2. Decision tree for establishing a surrogate buffer for enteric
maleate (pKa 5.8), citrate (pKa 5.7), succinate (pKa 5.2), coated products. Adapted from Blechar et al (79) under the Creative
Commons Attribution License
and acetate (pKa 4.6) buffers to find a buffer species (https://creativecommons.org/licenses/by/4.0/).
that would promote similar dissolution as bicarbonate
(78). The time taken for 5% release (t5%) was used for
CONCLUSION
comparison because it is most representative of the
The evolution of media and buffers used in dissolution
coat dissolution as opposed to the whole dissolution
testing to achieve physiological relevance was reviewed.
profile. The observed trend of dissolution based on t5%
There are many important factors to be considered when
was that succinate matched BCB well for relatively fast
developing a biorelevant dissolution method such as pH,
dissolving formulations, and citrate was a good estimate
buffer species, buffer concentration, osmolality, viscosity,
for relatively slow dissolving ones. These media could be
surface tension, concentration and type of bile salts,
used as good starting points. Based on these findings,
lipolysis products, as well as physiological state, such as
the authors proposed a decision tree for establishing a
fasted and fed states. Physiologically relevant methods
surrogate buffer (Fig. 2).
usually do not apply compendial conditions and its use is
A physiologically relevant approach is of primary most meaningful in the development phase, rather than
importance not only to predict the in vivo performance in a QC environment for batch release, for example. At the
of a formulation under development, but also to assess same time, the information retrieved from such methods
the similarity of reference and test formulations in a can be used to set specifications for QC and regulatory
bioequivalence (BE) study. Our group assessed clinical purposes. One of the majors disconnects between
data of a failed BE study for EC pantoprazole tablets the in vivo environment and in vitro conditions is the
(83). The formulations used in the dissolution study buffer species and concentration. The human intestinal
were from the same batch as those used in the BE study. lumen is buffered by bicarbonate at low molarities, but
Both formulations complied with the USP specifications highly concentrated phosphate buffers are often used
and had a somewhat similar performance in phosphate in dissolution testing, which can give misleading results
buffer, but when tested in vivo they did were not during the drug product development phase. This is

MAY 2022 71
www.dissolutiontech.com
 especially true for delayed release (EC) formulations. bioequivalence and Biopharmaceutics Classification System. Eur.
For these formulations, using bicarbonate-based buffers J. Pharm. Biopharm. 2000, 50 (1), 3–12. DOI: 10.1016/S0939-
would be the most ideal in terms of physiological 6411(00)00091-6.
relevance. On the other hand, a biorelevant test will not 7. Sheng, J. J.; McNamara, D. P.; Amidon, G. L. Toward an in vivo
necessarily be a clinically relevant dissolution test. The dissolution methodology: a comparison of phosphate and
pragmatical hurdles of using BCB makes it desirous to bicarbonate buffers. Mol. Pharm. 2009, 6 (1), 29–39. DOI:
develop a surrogate method with simpler buffer systems. 10.1021/mp800148u.
This can be achieved on a case-by-case study by comparing 8. Wu, C.-Y.; Benet, L. Z. Predicting drug disposition via
the drug flux in BCB and other buffer solutions (which are application of bcs: transport/absorption/elimination interplay
often more diluted systems compared to compendial and development of a biopharmaceutics drug disposition
buffers). Precise mechanistic understanding of the in vivo classification system. Pharm. Res. 2005, 22 (1), 11–23.DOI:
and in vitro dissolution processes is imperative to provide 10.1007/s11095-004-9004-4.
physiological relevance for the dissolution method. Using 9. Dorożyński, P. P.; Kulinowski, P.; Mendyk, A.; Młynarczyk, A.;
such conditions can aid the drug product development Jachowicz, R. Novel application of MRI technique combined
process by increasing the likelihood to select formulations with flow-through cell dissolution apparatus as supportive
with improved in vivo performance. discriminatory test for evaluation of controlled release
formulations. AAPS PharmSciTech 2010, 11 (2), 588–597. DOI:
FUNDING 10.1208/s12249-010-9418-8.
Daniela Amaral Silva is supported by the Alberta Innovates 10. Boetker, J. P.; Rantanen, J.; Rades, T.; Müllertz, A.; Østergaard, J.;
Graduate Student Scholarship. Opinions, interpretations, Jensen, H. A new approach to dissolution testing by uv imaging
conclusion, and recommendations are those of the author and finite element simulations. Pharm. Res. 2013, 30 (5), 1328–
and not necessarily endorsed by the funding agency. 1337. DOI: 10.1007/s11095-013-0972-0.
11. Reppas, C.; Friedel, H.-D.; Barker, A. R.; Buhse, L. F.; Cecil, T. L.;
CONFLICT OF INTEREST Keitel, S.; Kraemer, J.; Morris, J. M.; Shah, V. P.; Stickelmeyer, M.
The authors declare that there is no conflict of interest P.; Yomota, C.; Brown, C. K. Biorelevant in vitro performance
regarding the publication of this article. testing of orally administered dosage forms—workshop report.
Pharm. Res. 2014, 31 (7), 1867–1876. DOI: 10.1007/s11095-014-
REFERENCES 1348-9.
1. Grady, H.; Elder, D.; Webster, G. K.; Mao, Y.; Lin, Y.; Flanagan, 12. Jantratid, E.; Janssen, N.; Reppas, C.; Dressman, J. B.
T.; Mann, J.; Blanchard, A.; Cohen, M. J.; Lin, J.; Kesisoglou, F.; Dissolution media simulating conditions in the proximal human
Hermans, A.; Abend, A.; Zhang, L.; Curran, D. Industry’s view gastrointestinal tract: an update. Pharm. Res. 2008, 25 (7), 1663.
on using quality control, biorelevant, and clinically relevant DOI: 10.1007/s11095-008-9569-4.
dissolution tests for pharmaceutical development, registration, 13. Azarmi, S.; Roa, W.; Löbenberg, R. Current perspectives in
and commercialization. J. Pharm. Sci. 2018, 107 (1), 34–41. DOI: dissolution testing of conventional and novel dosage forms. Int. J.
10.1016/j.xphs.2017.10.019. Pharm. 2007, 328, 12–21. DOI: 10.1016/j.ijpharm.2006.10.001.
2. Anand, O.; Yu, L. X.; Conner, D. P.; Davit, B. M. Dissolution testing 14. Fotaki, N.; Vertzoni, M. Biorelevant dissolution methods
for generic drugs: An FDA perspective. AAPS J. 2011, 13 (3), 328– and their applications in in vitro in vivo correlations for oral
335. DOI: 10.1208/s12248-011-9272-y. formulations. Open Drug Deliv J. 2010, 2 (2), 2–13. DOI:
3. Marroum, P. J. History and evolution of the dissolution 10.2174/1874126601004020002.
test. Dissolut. Technol. 2014, 21 (3), 11–16. DOI: 10.14227/ 15. Amaral Silva, D.; Al-Gousous, J.; Davies, N. M.; Bou Chacra, N.;
DT210314P11. Webster, G. K.; Lipka, E.; Amidon, G. L.; Löbenberg, R. Biphasic
4. Amidon, G. L.; Lennernäs, H.; Shah, V. P.; Crison, J. R. A dissolution as an exploratory method during early drug product
theoretical basis for a biopharmaceutic drug classification: development. Pharmaceutics 2020, 12 (5), 420. DOI: 10.3390/
the correlation of in vitro drug product dissolution and in pharmaceutics12050420.
vivo bioavailability. Pharm. Res. 1995, 12 (3), 413–420. DOI: 16. Amaral Silva, D.; Al-Gousous, J.; Davies, N. M.; Bou Chacra, N.;
10.1023/A:1016212804288. Webster, G. K.; Lipka, E.; Amidon, G.; Löbenberg, R. Simulated,
5. Bou-Chacra, N.; Melo, K. J. C.; Morales, I. A. C.; Stippler, E. biorelevant, clinically relevant or physiologically relevant
S.; Kesisoglou, F.; Yazdanian, M.; Löbenberg, R. Evolution of dissolution media: the hidden role of bicarbonate buffer. Eur.
choice of solubility and dissolution media after two decades of J. Pharm. Biopharm. 2019, 142 (May), 8–19. DOI: 10.1016/j.
Biopharmaceutical Classification System. AAPS J. 2017, 19 (4), ejpb.2019.06.006.
989–1001. DOI: 10.1208/s12248-017-0085-5. 17. Mudie, D. M.; Samiei, N.; Marshall, D. J.; Amidon, G. E.;
6. Löbenberg, R.; Amidon, G. L. New Scientific approaches to Bergström, C. A. S. Selection of in vivo predictive dissolution
international regulatory standards. Modern bioavailability,

72 MAY 2022
www.dissolutiontech.com
 media using drug substance and physiological properties. AAPS Pharmacokinet. 1998, 23 (2), 97–102. DOI: 10.1007/
J. 2020, 22 (2), 34. DOI: 10.1208/s12248-020-0417-8. BF03189322.
18. Fuchs, A.; Leigh, M.; Kloefer, B.; Dressman, J. B. Advances in the 29. Hörter, D.; Dressman, J. B. Influence of physicochemical
design of fasted state simulating intestinal fluids: FaSSIF-V3. properties on dissolution of drugs in the gastrointestinal tract.
Eur. J. Pharm. Biopharm. 2015, 94, 229–240. DOI: 10.1016/j. Adv. Drug Deliv. Rev. 2001, 46 (1–3), 75–87. DOI: 10.1016/s0169-
ejpb.2015.05.015. 409X(00)00130-7.
19. Vertzoni, M.; Dressman, J.; Butler, J.; Hempenstall, J.; Reppas, 30. Dressman, J. B.; Amidon, G. L.; Reppas, C.; Shah, V. P. Dissolution
C. Simulation of fasting gastric conditions and its importance testing as a prognostic tool for oral drug absorption: immediate
for the in vivo dissolution of lipophilic compounds. Eur. J. release dosage forms. Pharm. Res. 1998, 15 (1), 11–22. DOI:
Pharm. Biopharm. 2005, 60 (3), 413–417. DOI: 10.1016/j. 10.1023/A:1011984216775.
ejpb.2005.03.002. 31. Mudie, D. M.; Amidon, G. L.; Amidon, G. E. Physiological
20. Galia, E.; Nicolaides, E.; Hörter, D.; Löbenberg, R.; Reppas, C.; parameters for oral delivery and in vitro testing. Mol. Pharm.
Dressman, J. B. Evaluation of various disssolution media for 2010, 7 (5), 1388–1405. DOI: 10.1021/mp100149j.
predicting in vivo performance of class I and II drugs. Pharm. Res. 32. Kalantzi, L.; Goumas, K.; Kalioras, V.; Abrahamsson, B.;
1998, 15 (5), 698–705. DOI: 10.1023/A:1011910801212. Dressman, J. B.; Reppas, C. Characterization of the human
21. Fuchs, A.; Dressman, J. B. Composition and physicochemical upper gastrointestinal contents under conditions simulating
properties of fasted-state human duodenal and jejunal fluid: a bioavailability/bioequivalence studies. Pharm. Res. 2006, 23 (1),
critical evaluation of the available data. J. Pharm. Sci. 2014, 103 165–176. DOI: 10.1007/s11095-005-8476-1.
(11), 3398–3411. DOI: 10.1002/jps.24183. 33. Gibaldi, M.; Feldman, S. Mechanisms of surfactant effects on
22. Hens, B.; Tsume, Y.; Bermejo, M.; Paixao, P.; Koenigsknecht, drug absorption. J. Pharm. Sci. 1970, 59 (5), 579–589. DOI:
M. J.; Baker, J. R.; Hasler, W. L.; Lionberger, R.; Fan, J.; Dickens, 10.1002/jps.2600590502.
J.; Shedden, K.; Wen, B.; Wysocki, J.; Loebenberg, R.; Lee, A.; 34. Ruby, M. V; Davis, A.; Schoof, R.; Eberle, S.; Sellstone, C.
Frances, A.; Amidon, G.; Yu, A.; Benninghoff, G.; Salehi, N.; M. Estimation of lead and arsenic bioavailability using a
Talattof, A.; Sun, D.; Amidon, G. L. Low buffer capacity and physiologically based extraction test. Environ. Sci. Technol. 1996,
alternating motility along the human gastrointestinal tract: 30 (2), 422–430. DOI: 10.1021/es950057z.
implications for in vivo dissolution and absorption of ionizable 35. <711> Dissolution. United States Pharmacopeia and National
drugs. Mol. Pharm. 2017, 14 (12), 4281–4294. DOI: 10.1021/acs. Formulary US P43–NF 38 2S; The United States Pharmacopeial
molpharmaceut.7b00426. Convention, Inc.: Rockville, MD, 2020.
23. Koenigsknecht, M.; Baker, J.; Wen, B.; Frances, A.; Zhang, H.; 36. Vertzoni, M.; Pastelli, E.; Psachoulias, D.; Kalantzi, L.; Reppas, C.
Yu, A.; Zhao, T.; Tsume, Y.; Pai, M. P.; Bleske, B. E.; Zhang, X.; Estimation of intragastric solubility of drugs : in what medium?
Lionberger, R.; Lee, A.; Amidon, G. L.; Hasler, W. L.; Xun, D.. In Pharm. Res. 2007, 24 (5), 909–917. DOI: 10.1007/s11095-006-
vivo dissolution and systemic absorption of immediate release 9209-9.
ibuprofen in human gastrointestinal tract under fed and fasted 37. Macheras, P.; Koupparis, M.; Tsaprounis, C. Drug dissolution
conditions. Mol. Pharm. 2017, 14 (12), 4295–4304. DOI: studies in milk using the automated flow injection serial dynamic
10.1021/acs.molpharmaceut.7b00425. dialysis technique. Int. J. Pharm. 1986, 33 (1–3), 125–136. DOI:
24. Riethorst, D.; Mols, R.; Duchateau, G.; Tack, J.; Brouwers, J.; 10.1016/0378-5173(86)90046-3.
Augustijns, P. Characterization of human duodenal fluids in 38. Nicolaides, E.; Galia, E.; Efthymiopoulos, C.; Dressman, J. B.;
fasted and fed state conditions. J. Pharm. Sci. 2016, 105 (2), 673– Reppas, C. Forecasting the in vivo performance of four low
681. DOI: 10.1002/jps.24603. solubility drugs from their in vitro dissolution data. Pharm. Res.
25. Markopoulos, C.; Andreas, C. J.; Vertzoni, M.; Dressman, J.; 1999, 16 (12), 1876–1882. DOI: 10.1023/A:1018959511323.
Reppas, C. In-vitro simulation of luminal conditions for evaluation 39. Kalantzi, L.; Persson, E.; Polentarutti, B.; Abrahamsson, B.;
of performance of oral drug products: choosing the appropriate Goumas, K.; Dressman, J. B.; Reppas, C. Canine intestinal
test media. Eur. J. Pharm. Biopharm. 2015, 93, 173–182. DOI: contents vs. simulated media for the assessment of solubility of
10.1016/j.ejpb.2015.03.009. two weak bases in the human small intestinal contents. Pharm.
26. Lambert, R.; Martin, F.; Vagne, M. Relationship between hydrogen Res. 2006, 23 (6), 1373–1381. DOI: 10.1007/s11095-006-0207-8.
ion and pepsin concentration in human gastric secretion. 40. Dissolution Testing of Immediate Release Solid Oral Dosage
Digestion 1968, 1 (2), 65–77. DOI: 10.1159/000196835. Forms; Guidance for Industry; U.S. Department of Health and
27. Dressman, J. B.; Berardi, R. R.; Dermentzoglou, L. C.; Russell, Human Services, Food and Drug Administration, Center for
T. L.; Schmaltz, S. P.; Barnett, J. L.; Jarvenpaa, K. M. Upper Drug Evaluation and Research (CDER), U.S. Government Printing
gastrointestinal (GI) pH in young, healthy men and women. Pharm. Office: Washington, DC, 1997.
Res. 1990, 7 (7), 756–761. DOI: 10.1023/a:1015827908309. 41. McConnell, E. L.; Fadda, H. M.; Basit, A. W. Gut instincts:
28. Efentakis, M.; Dressman, J. B. Gastric juice as a dissolution explorations in intestinal physiology and drug delivery.
medium: surface tension and PH. Eur. J. Drug Metab.
MAY 2022 73
www.dissolutiontech.com
 Int. J. Pharm. 2008, 364 (2), 213–226. DOI: 10.1016/j. concept and possible in vitro-in vivo correlation. Biol. Pharm.
ijpharm.2008.05.012. Bull. 2004, 27 (10), 1630–1635. DOI: 10.1248/bpb.27.1630.
42. Persson, E. M.; Gustafsson, A.-S.; Carlsson, A. S.; Nilsson, R. G.; 54. Kossena, G. A.; Charman, W. N.; Boyd, B. J.; Dunstan, D. E.; Porter,
Knutson, L.; Forsell, P.; Hanisch, G.; Lennernäs, H.; Abrahamsson, C. J. H. Probing drug solubilization patterns in the gastrointestinal
B. The effects of food on the dissolution of poorly soluble drugs tract after administration of lipid‐based delivery systems: a
in human and in model small intestinal fluids. Pharm. Res. 2005, phase diagram approach. J. Pharm. Sci. 2004, 93 (2), 332–348.
22 (12), 2141–2151. DOI: 10.1007/s11095-005-8192-x. DOI: 10.1002/jps.10554.
43. Wei, H.; Löbenberg, R. Biorelevant dissolution media as a 55. Jede, C.; Wagner, C.; Kubas, H.; Weigandt, M.; Weber, C.;
predictive tool for glyburide a class II Drug. Eur. J. Pharm. Sci. Lecomte, M.; Badolo, L.; Koziolek, M.; Weitschies, W. Improved
2006, 29 (1), 45–52. DOI: 10.1016/j.ejps.2006.05.004. prediction of in vivo supersaturation and precipitation of poorly
44. Psachoulias, D.; Vertzoni, M.; Butler, J.; Busby, D.; Symillides, M.; soluble weakly basic drugs using a biorelevant bicarbonate
Dressman, J.; Reppas, C. An in vitro methodology for forecasting buffer in a gastrointestinal transfer model. Mol. Pharm. 2019,
luminal concentrations and precipitation of highly permeable 16 (9), 3938–3947. DOI: 10.1021/acs.molpharmaceut.9b00534.
lipophilic weak bases in the fasted upper small intestine. Pharm. 56. Litou, C.; Vertzoni, M.; Xu, W.; Kesisoglou, F.; Reppas, C. The
Res. 2012, 29 (12), 3486–3498. DOI: 10.1007/s11095-012- impact of reduced gastric acid secretion on dissolution of salts
0844-z. of weak bases in the fasted upper gastrointestinal lumen: data
45. Cristofoletti, R.; Dressman, J. B. Matching phosphate and maleate in biorelevant media and in human aspirates. Eur. J. Pharm.
buffer systems for dissolution of weak acids: equivalence in Biopharm. 2017, 115, 94–101. DOI: 10.1016/j.ejpb.2017.02.009.
terms of buffer capacity of bulk solution or surface pH? Eur. 57. Boni, J. E.; Brickl, R. S.; Dressman, J. Is Bicarbonate buffer suitable
J. Pharm. Biopharm. 2016, 103, 104–108. DOI: 10.1016/j. as a dissolution medium? J. Pharm. Pharmacol. 2007, 59 (10),
ejpb.2016.03.024. 1375–1382. DOI: 10.1211/jpp.59.10.0007.
46. Cristofoletti, R.; Dressman, J. B. FaSSIF-V3, but not compendial 58. Tsume, Y.; Mudie, D. M.; Langguth, P.; Amidon, G. E.; Amidon, G.
media, appropriately detects differences in the peak and L. The Biopharmaceutics Classification System: subclasses for in
extent of exposure between reference and test formulations of vivo predictive dissolution (IPD) methodology and IVIVC. Eur. J.
ibuprofen. Eur. J. Pharm. Biopharm. 2016, 105, 134–140. DOI: Pharm. Sci. 2014, 57, 152–163. DOI: 10.1016/j.ejps.2014.01.009.
10.1016/j.ejpb.2016.06.004. 59. Allen, A.; Flemström, G. Gastroduodenal mucus bicarbonate
47. Söderlind, E.; Karlsson, E.; Carlsson, A.; Kong, R.; Lenz, A.; barrier: protection against acid and pepsin. Am. J. Physiol. Cell
Lindborg, S.; Sheng, J. J. Simulating fasted human intestinal Physiol. 2005, 288 (1), C1–C19. DOI: 10.1152/ajpcell.00102.2004.
fluids: understanding the roles of lecithin and bile acids. Mol. 60. Scott, N.; Patel, K.; Sithole, T.; Xenofontos, K.; Mohylyuk, V.; Liu,
Pharm. 2010, 7 (5), 1498–1507. DOI: 10.1021/mp100144v. F. Regulating the pH of bicarbonate solutions without purging
48. Fagerberg, J. H.; Tsinman, O.; Sun, N.; Tsinman, K.; Avdeef, A.; gases: application to dissolution testing of enteric coated tablets,
Bergström, C. A. S. Dissolution rate and apparent solubility pellets and microparticles. Int. J. Pharm. 2020, 585, 119562. DOI:
of poorly soluble drugs in biorelevant dissolution media. Mol. 10.1016/j.ijpharm.2020.119562.
Pharm. 2010, 7 (5), 1419–1430. DOI: 10.1021/mp100049m. 61. Fadda, H. M.; Merchant, H. A.; Arafat, B. T.; Basit, A. W.
49. Klein, S. The use of biorelevant dissolution media to forecast the Physiological bicarbonate buffers: stabilisation and use as
in vivo performance of a drug. AAPS J. 2010, 12 (3), 397–406. dissolution media for modified release systems. Int. J. Pharm.
DOI: 10.1208/s12248-010-9203-3. 2009, 382 (1–2), 56–60. DOI: 10.1016/j.ijpharm.2009.08.003.
50. Dressman, J. B.; Reppas, C. In vitro–in vivo correlations for 62. Liu, F.; Merchant, H. A.; Kulkarni, R. P.; Alkademi, M.; Basit, A. W.
lipophilic, poorly water-soluble drugs. Eur. J. Pharm. Sci. 2000, Evolution of a physiological pH 6.8 bicarbonate buffer system:
11 (Suppl. 2), S73–S80. DOI: 10.1016/S0928-0987(00)00181-0. application to the dissolution testing of enteric coated products.
51. Nicolaides, E.; Symillides, M.; Dressman, J. B.; Reppas, C. Eur. J. Pharm. Biopharm. 2011, 78 (1), 151–157. DOI: 10.1016/j.
Biorelevant dissolution testing to predict the plasma profile of ejpb.2011.01.001.
lipophilic drugs after oral administration. Pharm. Res. 2001, 18 63. Sakamoto, A.; Izutsu, K.; Yoshida, H.; Abe, Y.; Inoue, D.; Sugano, K.
(3), 380–388. DOI: 10.1023/A:1011071401306. Simple bicarbonate buffer system for dissolution testing: floating
52. Lue, B.-M.; Nielsen, F. S.; Magnussen, T.; Schou, H. M.; Kristensen, lid method and its application to colonic drug delivery system.
K.; Jacobsen, L. O.; Müllertz, A. Using biorelevant dissolution to J. Drug Deliv. Sci. Technol. 2021, 63, 102447. DOI: 10.1016/j.
obtain ivivc of solid dosage forms containing a poorly-soluble jddst.2021.102447.
model compound. Eur. J. Pharm. Biopharm. 2008, 69 (2), 648– 64. McNamara, D. P.; Whitney, K. M.; Goss, S. L. Use of a physiologic
657. DOI: 10.1016/j.ejpb.2007.12.013. bicarbonate buffer system for dissolution characterization of
53. Khan, M. Z. I.; Rausl, D.; Zanoski, R.; Zidar, S.; Mikulcić, J. H.; ionizable drugs. Pharm. Res. 2003, 20 (10), 1641–1646. DOI:
Krizmanić, L.; Eskinja, M.; Mildner, B.; Knezević, Z. Classification 10.1023/A:1026147620304.
of loratadine based on the biopharmaceutics drug classification 65. Goyanes, A.; Hatton, G. B.; Merchant, H. A.; Basit, A. W.

74 MAY 2022
www.dissolutiontech.com
 Gastrointestinal release behaviour of modified-release fluid boundary layer and the apparent effective pKa controlling
drug products: dynamic dissolution testing of mesalazine dissolution of acids and bases. Mol. Pharm. 2019, 16 (6), 2626–
formulations. Int. J. Pharm. 2015, 484 (1–2), 103–108. DOI: 2635. DOI: 10.1021/acs.molpharmaceut.9b00187.
10.1016/j.ijpharm.2015.02.051. 75. Salehi, N.; Al-Gousous, J.; Mudie, D. M.; Amidon, G. L.; Ziff, R. M.;
66. Garbacz, G.; Kołodziej, B.; Koziolek, M.; Weitschies, W.; Klein, Amidon, G. E. Hierarchical mass transfer analysis of drug particle
S. A dynamic system for the simulation of fasting luminal pH- dissolution, highlighting the hydrodynamics, pH, particle size, and
gradients using hydrogen carbonate buffers for dissolution buffer effects for the dissolution of ionizable and nonionizable
testing of ionisable compounds. Eur. J. Pharm. Sci. 2014, 51, drugs in a compendial dissolution vessel. Mol. Pharm. 2020, 17
224–231. DOI: 10.1016/j.ejps.2013.09.020. (10), 3870–3884. DOI: 10.1021/acs.molpharmaceut.0c00614.
67. Hofmann, M.; García, M. A.; Al-Gousous, J.; Ruiz-Picazo, A.; 76. Álvarez, C.; Núñez, I.; Torrado, J. J.; Gordon, J.; Potthast, H.;
Thieringer, F.; Nguyen, M. A.; Månsson, W.; Galle, P. R.; Langguth, García-Arieta, A. Investigation on the possibility of biowaivers
P. In vitro prediction of in vivo absorption of ibuprofen from for ibuprofen. J. Pharm. Sci. 2011, 100 (6), 2343–2349. DOI:
suspensions through rational choice of dissolution conditions. 10.1002/jps.22472.
Eur. J. Pharm. Biopharm. 2020, 149, 229–237. DOI: 10.1016/j. 77. Amaral Silva, D.; Davies, N. M.; Doschak, M. R.; Al-Gousous, J.;
ejpb.2020.02.009. Bou-Chacra, N.; Löbenberg, R. Mechanistic understanding of
68. Al-Gousous, J.; Ruan, H.; Blechar, J. A.; Sun, K. X.; Salehi, N.; underperforming enteric coated products: opportunities to add
Langguth, P.; Job, N. M.; Lipka, E.; Loebenberg, R.; Bermejo, clinical relevance to the dissolution test. J. Control. Release 2020,
M.; Amidon, G. E.; Amidon, G. L. Mechanistic analysis and 325, 323–334. DOI: 10.1016/j.jconrel.2020.06.031.
experimental verification of bicarbonate-controlled enteric 78. Blechar, J. A.; Al-Gousous, J.; Wilhelmy, C.; Postina, A. M.;
coat dissolution: potential in vivo implications. Eur. J. Pharm. Getto, M.; Langguth, P. Toward mechanistic design of surrogate
Biopharm. 2019, 139, 47–58. DOI: 10.1016/j.ejpb.2019.03.012. buffers for dissolution testing of pH-dependent drug delivery
69. Tsume, Y.; Langguth, P.; Garcia-Arietac, A.; Amidon, G. L. In silico systems. Pharmaceutics 2020, 12 (12), E1197.DOI: 10.3390/
prediction of drug dissolution and absorption with variation pharmaceutics12121197.
in intestinal ph for bcs class ii weak acid drugs: ibuprofen and 79. Aghabegi Moghanjoughi, A.; Khoshnevis, D.; Zarrabi, A. A
ketoprofen. Biopharm. Drug Dispos. 2012, 33 (7), 366–377. DOI: concise review on smart polymers for controlled drug release.
10.1002/bdd.1800. Drug Deliv. Transl. Res. 2016, 6 (3), 333–340. DOI: 10.1007/
70. Mooney, K. G.; Mintun, M. A.; Himmelstein, K. J.; Stella, V. J. s13346-015-0274-7.
Dissolution kinetics of carboxylic acids II: Effect of buffers. J. 80. Ibekwe, V. C.; Fadda, H. M.; McConnell, E. L.; Khela, M. K.; Evans,
Pharm. Sci. 1981, 70 (1), 22–32. DOI: 10.1002/jps.2600700104. D. F.; Basit, A. W. Interplay between intestinal pH, transit time
71. Krieg, B. J.; Taghavi, S. M.; Amidon, G. L.; Amidon, G. E. In vivo and feed status on the in vivo performance of pH responsive
predictive dissolution: comparing the effect of bicarbonate and ileo-colonic release systems. Pharm. Res. 2008, 25 (8), 1828–
phosphate buffer on the dissolution of weak acids and weak 1835. DOI: 10.1007/s11095-008-9580-9.
bases. J. Pharm. Sci. 2015, 104 (9), 2894–2904. DOI: 10.1002/ 81. Nguyen, D. A.; Fogler, H. S. Facilitated diffusion in the dissolution
jps.24460. of carboxylic polymers. AIChE J. 2005, 51 (2), 415–425. DOI:
72. Mooney, K. G.; Mintun, M. A.; Himmelstein, K. J.; Stella, V. J. 10.1002/aic.10329.
Dissolution kinetics of carboxylic acids i: effect of ph under 82. Sugano, K. Theoretical comparison of hydrodynamic diffusion
unbuffered conditions. J. Pharm. Sci. 1981, 70 (1), 13–22. DOI: layer models used for dissolution simulation in drug discovery
10.1002/jps.2600700103. and development. Int. J. Pharm. 2008, 363 (1–2), 73–77. DOI:
73. Krieg, B. J.; Taghavi, S. M.; Amidon, G. L.; Amidon, G. E. In vivo 10.1016/j.ijpharm.2008.07.002.
predictive dissolution: transport analysis of the co2, bicarbonate 83. Amaral Silva, D.; Gomes Davanço, M.; Davies, N. M.; Krämer, J.;
in vivo buffer system. J. Pharm. Sci. 2014, 103, 3473–3490. DOI: de Oliveira Carvalho, P.; Löbenberg, R. Physiologically relevant
10.1002/jps.24108. dissolution conditions towards improved in vitro - in vivo
74. Al-Gousous, J.; Salehi, N.; Amidon, G. E.; Ziff, R. M.; Langguth, relationship - A case study with enteric coated pantoprazole
P.; Amidon, G. L. Mass transport analysis of bicarbonate buffer: tablets. Int. J. Pharm. 2021, (10), 605. DOI: 10.1016/j.
effect of the CO2–H2CO3 hydration–dehydration kinetics in the ijpharm.2021.120857.

MAY 2022 75
www.dissolutiontech.com